Transition metal phosphide catalysts for hydrogen oxidation reaction Shamsul Izhar, Masatoshi Nagai * Graduate School of Bio-applications and Systems Engineering, Tokyo University of Agriculture and Technology, 2-24 Nakamachi, Koganei, Tokyo 184-8588, Japan 1. Introduction Fuel cells utilize hydrogen gas to generate clean and useful energy by the electrocatalytic oxidation of hydrogen (HOR). At present, the best and most practical catalyst for the HOR is the Pt- alloy catalysts. However, because these catalysts are expensive, current studies are concentrated on exploring a more economical catalyst based on non-noble materials, for example, carbides and perovskites. One of the non-noble catalysts considered as a potential candidate is the transition metal phosphide, which is generally known to demonstrate excellent activity for hydro- desulfurization (HDS) and hydrodenitrogenation (HDN) in the petroleum industries. In addition to its activity, these materials exhibit excellent physical and mechanical properties [1–5], for instance, electric conductivity, hardness, wear and corrosion resistance. Recently, monometallic phosphides have been studied for their use in electrode reactions, porous anode catalysis and in Li electrodes [6]. Daimon and Kurobe [7] reported that the addition of phosphorus to Pt–Ru resulted in a higher catalytic activity due to the effective particle size reduction of Pt–Ru. Phosphides have also recently been used for methanol oxidation in DMFCs [8]. In the present study, we investigated the HOR activities for some single and binary phosphide compounds. The phosphides were also investigated for their resistivity in acidic media due to the highly acidic environment of the PEFC. 2. Experimental 2.1. Catalyst preparation The precursors of the metal phosphides (WP, CoP or NiP) were prepared by combining stoichiometric quantities of ammonium metatungstate ((NH 4 ) 6 H 2 W 12 O 40 ÁnH 2 O), cobalt nitrate (Co(N- O 3 ) 2 ÁnH 2 O) or nickel nitrate (Ni(NO 3 ) 2 ÁnH 2 O) with an aqueous solution of diammonium phosphate ((NH 4 ) 2 HPO 4 ). For the bimetallic phosphides (Ni–WP and Co–WP), ammonium meta- tungstate or ammonium molybdate was added to cobalt nitrate, (Co(NO 3 ) 2 ÁnH 2 O) or nickel nitrate together with diammonium phosphate. As for the Co–MoP, the same starting material was used except that ammonium metatungstate was replaced with ammo- nium heptamolybdate ((NH 4 ) 6 Mo 7 O 24 ÁnH 2 O). The bimetallic ratios of Ni to W, Co to W and Co to Mo were set constant at 1:1. The solids were dissolved in water at 313 K with stirring and then dried at 393 K. The catalysts were ground before being calcined for 5 h at 773 K. These oxide precursors were then reduced at temperatures shown in Table 1 at a ramping rate of 1 K min À1 in a stream of H 2 and maintained at the final temperature for 2 h. Prior to phosphidation, temperature-programmed reduction (TPR) was carried out in a flow of hydrogen, and the evolution of water (m/ z = 18) was measured online using a quadrupole mass spectro- meter (Baltzer). The mixture was then cooled and passivated under a stream of 1% O 2 /He. The dissolution of the metal phosphides in acid was carried out by adding 2 mg of the catalyst to 10 mL of 0.5 M H 2 SO 4 , then followed by mixing in a tube mixer for 24 h prior to analysis by inductively coupled plasma spectroscopy (ICPS, Shimadzu). Catalysis Today 146 (2009) 172–176 ARTICLE INFO Article history: Available online 23 February 2009 Keywords: Transition metal phosphides Hydrogen oxidation reaction Phosphide dissolution in acid ABSTRACT A series of transition metal phosphides consisting of tungsten (WP), cobalt (CoP), nickel (NiP), nickel– tungsten (Ni–WP), cobalt–tungsten (Co–WP) and cobalt–molybdenum (Co–MoP) was prepared by a temperature-programmed reduction process. Hydrogen oxidation activities for the phosphide compounds were studied. The hydrogen oxidation reaction (HOR) was determined using a half-cell rotating disc electrode in a 0.5 M H 2 SO 4 electrolyte. In addition, the passivity of the phosphides in an acidic environment was studied by ICP. The Co–WP demonstrated the highest activity for the HOR based on chronoamperometry and the exchange current density. The single WP exhibited a low HOR activity, but the incorporation of Ni or Co enhanced the activity. In contrast, WP exhibited a high passivity, while NiP and CoP dissolved in the 0.5 M H 2 SO 4 . The high activities over Co–WP and Ni–WP were attributed to the role of WP in preventing the Co and Ni from corroding in the acidic media. ß 2009 Elsevier B.V. All rights reserved. * Corresponding author. Tel.: +81 42 388 7060; fax: +81 42 388 7060. E-mail address: mnagai@cc.tuat.ac.jp (M. Nagai). Contents lists available at ScienceDirect Catalysis Today journal homepage: www.elsevier.com/locate/cattod 0920-5861/$ – see front matter ß 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.cattod.2009.01.036